The Man Who Mastered Al: The Fastest Way To Master & Use AI Agents In Your Business, Content & Life

Paul Lipsky 0:00 Some of these tools I'm gonna show you are really easy to use. Think that a lot of people sleep on what's available to you through Google, and people just like get hyper-focused on ChatGPT or Claude without realizing that these incredible tools exist. And they have something called avatars. So when you set this up for the first time, you scan your face, and then you have a digital avatar of yourself that you can use for any video. You can't do that with Claude. Just learn this stuff and then bring it to businesses who don't know how to do it. That feels so easy to me.

Calum Johnson 0:29 For someone that's listening to this conversation, what would you have as like the immediate step one?

Paul Lipsky 0:33 It's really not that hard. I would actually start with no-

Calum Johnson 0:39 Okay. Today's guest is Paul Lipsky, and before we even get into it, I just have to come out and say it. I did not expect to record this episode on this topic, and the reason that I say that is that a few weeks ago, I have a call with Paul, and during that call he starts to share with me all of these different tools and workflows that he's using. And then he gets to this part where he starts showing me these AI tools that I've never really heard anyone talk about. I'm not talking about Claude or the new ChatGPT feature. I'm talking about Google suite of AI tools, specifically Google AI Studio 2.0, Gemini, and Gemini Notebook. And when he showed me these workflows, I actually interrupted him on the call and I told him, I said, "You know what? Let's just focus the whole episode around showing people how they can use those three tools in their business, their content, and their life." And so that's what we did. And so by the end of this conversation, you're gonna learn how to use your Gmail better. You're gonna see how you can build apps and even create content and do research in ways that you never expected. Specifically what Paul showed with Gemini Notebook was mind-blowing to me. It literally changed the way that I do research and preparation for episodes. And so you know what? I'm not gonna say anything more. Let's just get into the conversation. You're listening to The Calum Johnson Show with Paul Lipsky. I've seen so many videos online about Claude and ChatGPT, but one of the unique things for you is that a lot of your most popular videos are actually about Google's AI tools.

Calum Johnson 2:16 And so if someone has clicked on this video and they've never used any of those tools, what will be the value to them and in their lives and in their businesses of learning about those tools?

Paul Lipsky 2:28 I think people are gonna be pretty surprised at what Google has to offer because it really, um, spreads across the entire spectrum. You can start as a complete beginner with their tools and get a ton of value out of them, and have a lot of fun using them too. But even for the more advanced users, you can do a whole lot with the Google AI suite of products that they have. And I think one thing that makes it so unique is that it does en- encapsulate so many different tools that you get access to through Google AI. It's pretty neat.

Calum Johnson 3:00 Yeah. You know, one of the ... I think your background is so interesting, Paul, and one of the things, 'cause we spoke before this, and one of the things that you mentioned, um, and it's also in your YouTube bio, is you say, "Unlike a lot of other YouTubers who talk about AI, I'm not a programmer or a coder. I'm just a regular guy who likes to use AI in my personal life and to streamline my businesses." And so I'm curious for you to kind of like take us back in time for a second and talk to us about being a regular guy. But also, what is that first moment, like that light bulb moment when you use AI and you see the power of this technology for the first time?

Paul Lipsky 3:44 Yeah. Yeah. So, um, quick background on me. I had a pretty ordinary kind of, uh, career path at first. Uh, I went, actually went to law school. I was a lawyer for a while. But I wasn't really happy doing that, and so I started this side gig selling some stuff on eBay, and it actually took off. I ended up doing more, making more money from that than being a lawyer. So I went full time with that, started documenting what I was doing on YouTube, and so that's where I got started on YouTube, was, was telling people about eBay and Amazon and Facebook Marketplace and these other places I was selling. And then somewhere around 2025, that year, I noticed this big shift where the AI tools that were out there-- I'd tried them before and honestly I wasn't very impressed with them at all. But in 2025 there was a big shift, and suddenly the tools became a lot more accessible and helpful for people like me who doesn't have a background in coding, isn't a developer, and just wants help with like, at that time it was just like my emails and my sales copy that I wanted help with. And so I think it was probably around spring 2025, I signed up for Claude, I paid the $20 for it, and I was blown away. I was like, "This is incredible what it's able to do, how it's able to help me organize all of my projects and write sales copy for me." I was completely blown away. And then from that point on, I pretty much went all in with it, 'cause I really saw like if this is just the beginning of what we're experiencing, this has huge, huge potential.

Calum Johnson 5:17 Yeah. You know, I, I actually think it's so, uh, relatable, especially for people right now. Like I think, I think to myself that a lot of people are going through that moment right now, specifically with Claude, of like that shift of seeing what's possible and how it can help you, not only in your life, but like business, career. Talk to me specifically about the Google AI tools and, and when did those come into the picture? Because a lot of people, it's like it starts with ChatGPT or Claude, and it hasn't even gone past that. Like what even made you curious about understanding what Google was doing in the AI space?

Paul Lipsky 5:58 So I do feel like Google was a little bit late. I think they were kind of caught off guard by how quickly ChatGPT and Claude kind of took off. But- Once they figured it out, they had a huge advantage because we already use so many Google products, Search, Gmail, Calendar, and, or Drive. And what's awesome about what Google's doing is that they are bringing AI into the products that we already use. And so the integration with AI into them and between those different products is so seamless in a way that really no one else can do it. And so when I saw them start to roll it out, I got really excited because I don't wanna have to redo everything I- I've set up in my business. You know, find a new email provider and set up a new calendar. I wanna use the tools I already love. And the fact that Google can now integrate with all that and, and connect it all with AI is pretty incredible, and I think that's where people are gonna see a lot of value out of it.

Calum Johnson 7:01 Yeah. You know, a- and, and it's an interesting point that you bring up because Google should naturally have an advantage in this space. Like, we- we're already so aware of them. They're already so in our lives, even without us really knowing. Why do you think ... Because for most people, if you look at the data, when it comes to AI, the majority of Americans have used ChatGPT. In second, in kind of like a smaller way, people have used Claude. Why do you think that ChatGPT and Claude demon- like, dominate so much of the conversation, and then when we go to Gemini or, like, the Google suite of tools, it's not as ubiquitous, it's not as known?

Paul Lipsky 7:45 Well, I think they are catching up, or at least, or at least pulling in more of the market than they've had in the past. But I think a lot of it just comes down to that, like, first-to-market bias. Y- you know, the word ChatGPT, it's almost synonymous now with AI, and people say, "Let me ask Chat for that." That's what I'm hearing young people say now. And I'm not 100% sure if, if that applies just to ChatGPT or, like, all of them, kind of like when we say, "Let's Xerox something," right? But I think it's, a lot of it is that. It's just when people think of AI, they think of ChatGPT. But I think once people realize that there are other tools and start to test them, they're gonna find one that works better for them maybe than ChatGPT.

Calum Johnson 8:26 Yeah. You know, you know, one of the things that you mentioned, uh, earlier, Paul, you were talking about, uh, being like a regular guy, and you have this, uh, job. You're a lawyer, like you're an attorney.

Paul Lipsky 8:39 Mm-hmm.

Calum Johnson 8:39 And then within the space of a year or a couple of years, you're doing, uh, like, e-commerce and dropshipping, and now you're doing, like, content creation around AI. I think for a lot of people that are listening at home, that transition of going from, like, a 9-to-5 employee to you're making a full-time living off the internet and online is, like, something that people aspire to. Can you kind of just map out, like, what were those kind of key points in that journey, and then also how quickly it happened? Was that unexpected to you?

Paul Lipsky 9:21 Well, I'll answer that first, that last part first. It was very unexpected. [laughs] You know, I just didn't quite see it happening. But in many ways I shouldn't be surprised because it was something that I'd been seeking for a while and something I really had my eye on. And I think I'll just say, like, very broadly speaking, if you want to, you know, quit your job and make money online to have that freedom to be able to, like, travel and not be tied down to one location, there's always opportunities for that, and I think AI is opening up a ton of possibilities for that. It's just kind of being creative and seeing what's available and taking advantage of that. And that's what I saw in 2017 when I got started with, with e-commerce. I tried some other stuff in the past, some, some other e-commerce things. It didn't work out for me. And I saw some people doing this thing, dropshipping on eBay, and I gave it a shot. And once I saw that it worked, I was like, "Okay, I'm g- going all in with this," because i- you know, I don't wanna, um, you know, play around with other stuff. Like, if I find something that works, go all in with it. And I think that's part of the success is just kind of knowing that, like, when you see something that works, you gotta go for it 100% 'cause you don't know how long it's gonna work for, right? So that was, like, a very key moment, like when I made my first sale and just started working a lot more with eBay. Um, and then certainly starting my YouTube channel to be able to share and document what I was doing was really important because then as I pivoted over the years, you know, moving to Amazon, Facebook, TikTok, people were following me and coming with me on that journey. And now that I've transitioned to, uh, AI, people are coming along for that as well, so it's pretty cool to see.

Calum Johnson 11:03 Yeah. You know, you talk about, I- and I think about it even in my own story, that first moment of validation 'cause we, we start out online, and you're, you're kind of just trying something. You know? Like, you've, maybe you've seen someone's video or you've read something. You've kind of seen that it's possible, but until you experience it ... And so I'm curious for you to talk about almost, like, that, that first moment of, of real validation, like significant validation, whether it was, like, a dollar amount that you were seeing in your bank account. But, like, for you, when you look back, what was that moment when you were like, "Okay, this could actually be a full-time thing. It's not just, like, a fun side hustle or, like, weekend project. Like, I could actually do this"?

Paul Lipsky 11:47 Okay, I've never shared this before, so I'm gonna share it with you now. [laughs] You pulled it out of me. So yeah. So 2017 was the year I was growing that eBay business, and it, by the middle of the year it was doing quite well, but it was always sort of that feeling of, "Is this gonna last? Like, is this thing just gonna topple in a couple months? Or is this thing even really serious?" And then early 2018, I had one month Where I made more in one month than I was making all year as a lawyer. And that was kind of like a moment where I was like, "Yeah, I should quit my job now." [laughs] This is, this is, this is working. There's something here. And so that's when I decided to actually quit my job and go all in with the e-commerce and kind of have this s- uh, YouTube channel on the side with it.

Calum Johnson 12:36 Yeah. You know, it's, um, it's interesting because you mentioned that speaking today in 2026, there's even, like, more of an opportunity for that route, for that transition from, like, 9 to 5 to either being, like, a solopreneur or, like, building a business, um, specifically because of AI. And I know that you and I, we were speaking about this, um, even before we had the conversation and, like, we turned the cameras on, and you said this, you go, "A lot of people just don't know what they can do with AI." That's sort of the biggest roadblock right now, that people know it exists, they know it's useful, but what do you use it for? That's where most people are getting stuck with it. And so s- Paul, speaking specifically on these Google tools, 'cause I know we're gonna go through a few of them, can you just big picture, just give people the background and the context for these specific tools? So, like, Gemini, NotebookLM, the Google AI Studios. What can they use it for? Okay, so if you're someone who always has good ideas for the company that you work for or maybe even for your own business, I'm gonna share something that you're going to love. And so this is an episode marketing plan that I created with Gamma and then I presented it to our team. And so to give you guys the context, the reason I created this presentation was because I recently came up with a new way that I wanted us to promote each episode on social media. And I'm not a designer. And so what I would usually do is I would take all of the rough notes that I had for this strategy. I would try to organize it and flesh out the ideas on a Google document. Then I would have to format the document so that it looks half decent, and then I would send that document to my team and walk them through it step by step in our next meeting. And that entire process would take anywhere from a few hours to maybe even up to half a day of work. But with Gamma, I was able to just select this template that they have, paste in all of my rough bullet point notes, hit generate, and then get this beautiful looking presentation in 15 minutes. And the best part was just how easy it was to then change the designs and the images on the presentation. I just click on the Gamma agent here and tell it all of the changes that I want to make, and it did it for me. And so if you're listening at home and you want to create beautiful looking presentations in minutes, just go to that link in the description, try out Gamma today. Trust me, you won't regret it. Thank you to Gamma for sponsoring this episode. Let's get back to it.

Paul Lipsky 15:16 Yeah. Well, first I'll say that is the value gap right there, right? 'Cause so many people, businesses, business owners, they don't know how to use these tools. And if you can come in and figure that out, then you can be the person they go to. Some of these tools I'm gonna show you are really easy to use, but people don't figure them out. They're not gonna spend the time to figure them out, and so you can be that person for these companies. So I think there's actually a really big opportunity right now, uh, especially if-- for businesses owned by older people who just don't think they can figure it out. But the, the trick is that you can't let them know how easy it actually is. [laughs] Um, yeah, so I'm gonna show you some, some of these really cool tools. I think, you know, when people think about ChatGPT, right? So Google has their own version of it. It's called Gemini. That's their chatbot. So a lot of stuff that you know you can do with ChatGPT you can do with Gemini as well. But with a Google AI, uh, account, which is just your, your Google account, you also get access to all these other tools, and one of my favorites is NotebookLM. That's an incredible tool for research and for organizing data. And once you sort of graduate from those tools and want something a little bit more advanced, that comes with Google as well. They have something called Google AI Studio, which is used for, like, building out apps, building out websites. So there's a lot you can do with all of these tools.

Calum Johnson 16:45 Yeah. It's actually interesting because, um, I was thinking about this on the way here, that before you and I spoke in the conversation that we had about a week ago, I wasn't actually fully aware that all of these tools within Google even existed. Like, I was familiar with Gemini, and we've even seen it recently... Like, even when you do searches on Google, it's now using AI. I don't even know if it's using Gemini, but it's using AI to give you search results. Um, can you kind of just speak to, even just for the everyday person that's using G Suite right now, some of those kind of AI opportunities, almost like the, the hidden abilities that some of these G Suite tools n- or G Suite applications now have using AI and Gemini? Can you speak to that?

Paul Lipsky 17:37 Yeah. Yeah. So Google has put Gemini everywhere, so search is powered by Gemini. Um, and Gmail. So you go into Gmail, and now at the top, the search bar, that is powered by Gemini. Yeah, so inside of Gmail at the top, like, there's search now. It's powered by Gemini. And that's great [laughs] because I don't know if you've tried the search inside Gmail in the past, but it was pretty bad. So now you can actually ask questions like, uh, "What time was my po-- is my podcast recording today?" And it's not just going to give me the email that has the information, it's actually gonna answer it right there at the top for me. You need help writing an email, you just open up an email that you got, hit reply, and it can help you draft an email. I don't write emails anymore. I just type into the little bar at the bottom what I wanna say, like just garbled, and it just outputs for me a perfectly formatted email. So I've gotten really lazy with my emails because of that. Um, Google Sheets has some great tools as well to help you visualize data. It's just everywhere, and it just makes it so much easier to use it because it's where you already are.

Calum Johnson 18:48 Yeah. You know what? You know what, Paul? 'Cause this is, like, exciting. 'Cause I think we're gonna, uh ... Of course, we're gonna show people that wanna build businesses or solopreneurs some applications for these tools that they could be helpful in their workflows. But even just for everyday usage, I'm, like, curious and excited to show people how they could be getting more, uh, from their Gmail, from their Google Drive, from their Google Sheets. Um, so why don't you just share your screen, and we'll jump into it.

Paul Lipsky 19:18 This is Gemini. Yeah. This is Gemini. So this is, like, the ChatGPT for Google. So if you've used ChatGPT, anything that you use that for, you know, asking questions, you can come in here and just ask the questions in here. What is the weather today? You know, whatever it is, it's gonna behave very similar to ChatGPT. So if you're using an AI chatbot already, you're gonna just be able to do the same thing inside of Gemini. So one thing that's unique to Gemini is that it can create music for you. So you can just kind of put in here what you want the song to be about, and then it generates a soundtrack.

Calum Johnson 19:55 Yeah. Uh-huh, huh. Word up. Check the architect. My mind's the blueprint.

Paul Lipsky 19:59 Yeah. So just, that's just a quick sample of it. And you can create, like, in all these different styles. So if I come in here and click Create Music, you have, like, all these different styles. And, you know, it's not gonna win any awards, but it's a lot of fun to play around with, right? You can create images with it similar to what you do with ChatGPT, and it's really good at editing images. So here I had it create this one for me, and I didn't like that there were people in the background, so I said, "Here, remove the people from the background." Now we have the same exact image but without people in the background. So it's very good at image generation, and it's also very good at video generation. So ... Oh, this is a ... Where did the video go? Sorry. Yeah. So I have, like, a video here, and they have something called avatars. So when you set this up for the first time, you scan your face, and then you have a digital avatar of yourself that you can use for any video. I have to tell you about Google Gemini. It's their most capable AI model ever. It understands text, images, and even codes seamlessly. This is a total game changer- Yeah. So pretty good, right? So those are kind of, like, some of the unique, fun things you can do with it.

Calum Johnson 21:04 Just one quick thing on the video piece, 'cause that's very cool. So it scans your face. That video that you just played, you never recorded that?

Paul Lipsky 21:13 No. No. That came from one, from one scan of my face that I did three or two to three months ago.

Calum Johnson 21:20 Interesting. And the, and the different angles, 'cause it, it ... The, the camera shifted. It was showing different angles of you. That was also ... It just came up with that?

Paul Lipsky 21:29 Yeah. I mean, you can get very specific with your prompts. This one I literally just said, "Make a video about, from, of, of me," that's my avatar's name, "telling us about how awesome Google Gemini is." That's it. But you can get very specific with it and tell it to, like, change your clothes, change the background, and it can do all that stuff.

Calum Johnson 21:46 Interesting. And then when it comes to your voice, it also has your, it has your voice, like, saved?

Paul Lipsky 21:52 Yeah. It's pretty wild, right? [laughs] So when you're setting it up, you're, like, moving your face around, and you're also talking to it, and so it's capturing all of that and then duplicating it. And that's totally optional. I know some people aren't comfortable with that, so you don't have to use it.

Calum Johnson 22:07 Yeah. But it's one of the things ... 'Cause I, I even think about it in, like, a business use case, and I think about if you wanted to do, like, explainer videos or if you wanted to do ads, you know, where you're speaking to something. Or even, like, content, like talking head videos, but you didn't have time to film it or you didn't have the right setup. It's like you can ... The, the use cases kind of jump out.

Paul Lipsky 22:27 Yeah. Oh, yeah. And we're gonna get to that 'cause we're gonna ... We have some other tools that's gonna, that are gonna help with that as well.

Calum Johnson 22:32 Yeah. Yeah, that's super cool.

Paul Lipsky 22:34 Yeah. Um, yeah, so a lot of cool stuff with the images and the videos. It's also really good at creating infographics where you have text on images like this. So those are kind of the, like ... I wouldn't even call them basic, but you can come in here and kind of see just create image, create video, create music. So it's right there available for you.

Calum Johnson 22:56 Yeah. You know what? So, so I have, I have two questions. I'll, I'll start with the first one. On the image creation piece, 'cause I know Google obviously has NanoBanana. Is that, like, just seamlessly integrated in? Is that what's creating the images, or is this actually, like, a natively built feature within Gemini?

Paul Lipsky 23:16 No. This is, this is, uh, the latest NanoBanana. So right now when we're filming this, we have NanoBanana 2. So this is using NanoBanana 2. The, uh, the other images were creating NanoBanana 2. I lost them, but [laughs] whatever, wherever they are, that was using NanoBanana 2. Uh, so it's built in. You don't have to worry about, like, API. You can just tell it what image you wanna create, and it's gonna create it for you.

Calum Johnson 23:39 Yeah. That's interesting. You know what, Paul? 'Cause I think this is gonna be really valuable to people, is if I just take myself for a second, right? 'Cause I, I use G Suite. Like, I have a Gmail. I use Google Drive. I use Google Sheets. I use Google Documents. But I think that even ... I use those every day, but I'm not really using the AI features within it at all. If someone ... And even just for someone like me, what would you say is, like, the, the single almost, like, first thing to do or, like, most useful thing I could do with using AI within those tools? Like, what instantly comes to mind do you think would be valuable or the most valuable to people?

Paul Lipsky 24:27 Yeah. So I mentioned it before, but honestly, the integration with Gmail is, is fantastic. I can't show my Gmail because it's my Gmail. [laughs] It has, like, all my personal stuff in there. But it's so, so nice just using it to help you compose emails. Um, the other thing is because this is Google, you can come in here and just ask Gemini, "What are my latest emails?" And as long as it's the same account that you use for Gmail, it's gonna be able to pull in your emails. It can help you answer the emails right here from the chat. So all of that's good. Google Drive has some pretty interesting stuff as well. This is just, like, a quick example. Imagine you're, like, a footwear brand or you're looking to start one and you just wanted to do some research on the other competition out there. So I just asked Gemini to do some research on what the footwear market is like from the Gemini app, and it created for me this spreadsheet here, here with all this data on the different brands, where they're located, the CEOs, and the company revenue. So all of that's brought over into here by Gemini. And then from here, if you want to, you can click on Gemini on the right, and then you can, you can ask it to, to make changes here and do things. So I can say something like, uh, you know, "Sum it all up." I don't know. It's, like, a very s- silly example. But then it will sum it all up. Um, you can ask it to make changes to it. But my favorite tool, I'm trying to remember where it is. It's right, um, yeah, down here. You can click on Create a canvas. Oh, I have this up, that's why. Let me try it again. Create a canvas. And then, oh yeah, there it is. Sorry about that. So over here on the right, on the dropdown, it's here. It's called Canvas. And then you can say, "Turn this into a visual that I can look at." Something like that. Um, it's gonna take a minute, but it's gonna turn all this data into a more visual dashboard that's easier to look at. So we gotta give it a minute to do that, though.

Calum Johnson 26:33 Yeah. I, I'm curious for you, 'cause you've been, you've been using these tools for months. Are you getting to the point now where, like, even working within a Google Document or working within a spreadsheet, you're not really doing any, like, manual data entry? Like, you're not really just typing things in. You're just speaking to the AI, and it's doing the work. Like, how, how are you interfacing with it?

Paul Lipsky 26:58 Literally speaking to it. Literally speaking to it. So I use one of those, like, voice dictation apps, if you're familiar with those.

Calum Johnson 27:04 Like Whisperflow? Yeah. I do the same thing.

Paul Lipsky 27:06 Exactly.

Calum Johnson 27:06 Yeah.

Paul Lipsky 27:06 Whisperflow is the one I use. And so, like, for Gemini, I'll just come over to Gemini and, you know, they have a voice dictation button. So if you're, you don't have one, you could just click on this button right here, talk to it, and then it will send off that prompt and start working on it. And then inside of Google Drive, if I start a new document here, then down here at the bottom we have Write with Gemini. So I'll tell it, "Yeah, I'm putting together an outline for a video I'm working on. Can you go out and do the research for me, pull it in, and so I have, like, the first draft ready to go?" And by the way, this is that dashboard I was talking about. So it took all that data and just kind of put it into this nice-looking dashboard here.

Calum Johnson 27:47 Yeah. That's nice. That's nice. And I guess you could go back to it and you could make, like, changes if you wanted different-

Paul Lipsky 27:55 Yeah. You could just click on View data here to go back to the data or, yeah, you could just tell it in the chat, "You know, I don't like these colors. I want a different kind of graph." It will make all those changes.

Calum Johnson 28:04 Yeah. You know, one, one more question because I, I know we're gonna go to, um, NotebookLM, is when you're, um ... That feature within Gmail, 'cause I think that's something that people are gonna use, like, probably straight away after watching this, honestly, and I'm gonna use it. When you're asking it, you ... 'Cause you mentioned you don't actually draft emails anymore.

Paul Lipsky 28:28 Right.

Calum Johnson 28:28 It's the, it's the AI. Does that mean it actually has context, I guess, on, like, how you like to write emails, the structure, kind of, like, your tone of voice, the language that you use? Like, how good is that piece of it?

Paul Lipsky 28:44 Really good. Really good. So now when I open up my Gmail, again, I really wish I could show this to you, but [laughs] when I open up my Gmail, it actually pre-drafts the emails for me. N- not for all of them, but for a lot of them. And so I get a lot of emails from brands who wanna, like, sponsor videos on my channel, and it understands it. I didn, I never told it, I never told Gemini what my rates are for videos, but when I open up the drafts, it's in there because I've already sent emails to other brands with my rates. So it's already in there. And, uh, yeah, so, like, stuff like that. It's just, it just, it learns the more that you use it.

Calum Johnson 29:24 And it knows that what, from just, like, viewing your email inbox? Or is that something that you've actually trained it to do?

Paul Lipsky 29:31 Let me pop into Gemini because I want to show you this. So personal intelligence connects up all of your Google products. You come into Google, uh, connected apps right here. So we have Gmail, Google Docs, Keep, Drive, Calendar, Google Tasks. So what that does is it's pulling in context from all of those apps. So if I were to ask chat, um, "Tell me about my ... " Or, "What, when is my flight next month?" It's going to look through Gmail for emails from, from, uh, airlines, and it'll say, "Oh, this is when your flight is next month." Or if I say, "Put together an itinerary for the trip I'm taking next month," it's gonna pull in from multiple Gmail accounts. It's gonna look at my calendar. It's gonna look at maybe I've a Google Doc where I've already started planning out the trip, and it's gonna pull all of that together just sort of magically in the background.

Calum Johnson 30:25 Yeah. You know what, Paul? Before we, um- It's funny 'cause you mention, uh, like, the, the biggest roadblock. We, we spoke about this earlier. The biggest roadblock when it comes to using AI is people don't know what to use it for, and it's like, uh, there was so many of these features I just didn't even really know that Google could do it. Um, and I already am using G Suite, so it's just interesting. You know, one thing before we go into NotebookLM, um, I know right now with, with Claude and with ChatGPT, we're in a very interesting time, right? Like, uh, Fable obviously came out recently, got banned, and then has been released again. And then even with ChatGPT, uh, 5.6, uh, will be hitting the market soon. Probably when people have watched this, it probably would've hit the market. Can you just talk about what have you seen is the differences, um, between, like, Gemini's models and what it's capable of versus obviously Fable seems to have taken things to a new level, and then we'll see what GPT 5.6 does. Like, is it comparable?

Paul Lipsky 31:38 Yeah, it's a good question. So, uh, Fable and we don't-- I don't have access to 5.6. We're recording this, it's coming out literally tomorrow, so I'll know better tomorrow. But it does seem like it's gonna be, like those two are gonna be the, the best models in the world, better than what Gemini currently has out. But Google has this tendency to really cook in the background for a while when it comes to models. Their newest model is rumored to come out soon. I, I don't know when it's gonna come out, but the rumors are it's supposed to be very good and supposed to be a big jump up. Now, in general, though, what I will say is that if you start looking on X or even a lot of other YouTubers, they are going to obsess over what is the best model out there. But the truth is, for most people doing knowledge work, so things like emails, document writing, research, just everyday tasks, it's not going to make a huge difference which model you use. You may not notice a difference at all. So I like to focus a lot more on the tools themselves and the capabilities of them and just which one feels right for you, which one has the features that you want. Um, of course, they all have their strengths and weaknesses, and I think that for Gemini, some of those strengths are image generation is very good. Um, it's not quite as good as what ChatGPT has, but it, but is still very, very good and especially for creating infographics like this. It is one of the few tools that you can generate videos with. You can't do that with Claude. You can't do that with ChatGPT anymore. But Gemini does create videos, and it does it very well. This new model that I showed you before is excellent. It's very good at research. It has a huge context window, and I've heard that the new model's gonna have even a double of that context window. So it does have a lot of strengths and, and then, uh, yeah, and then just how it integrates with everything I think is another strength of it.

Calum Johnson 33:41 Yeah. You know what, just quickly on the, on the context window point, because I think that's actually heavily underrated. Like, w-we don't think about that. And it's funny 'cause we had a, I had a conversation, uh, on this show with, uh, Matt Wolfe a year ago. He's like another creator on YouTube, and this was in 2025. But at the time, he was talking about, like, the, the massive context window that Gemini has. And he said that, uh, "I use Gemini if I need to understand a massive research paper or a YouTube keynote, or I'll put in a three-hour transcript that I want broken into bullet points." Can you talk for a second from your vantage point some of the use cases of that massive context window and the deep research? Like, what's the value of that to people?

Paul Lipsky 34:32 Yeah. So I think all of those use cases that you just mentioned are spot on. Um, it's just anything where you just have so much data that you have to throw at it. So think you're a student, and you just have lots of notes from your semester that you have to study. You can dump them all into Gemini or NotebookLM, and it's going, it's going to be able to see all of that and actually use it in a helpful way for you. Um, I could see a lot of-- I know a lot of people in academia use it. A lot of researchers use it just because it, it, 'cause the context window is so big.

Calum Johnson 35:10 Yeah. You know what, Paul? Because, um, it's funny 'cause I mentioned this in the beginning. One of the most popular videos on your channel, I think it's the second most popular video on your channel, nearly half a million people have watched it, is on NotebookLM, which would be, like, so unexpected 'cause I think a lot of people haven't even heard of NotebookLM. And so if you take us to that moment, first of all, what inspires you to create that video on NotebookLM? And then why do you believe that it's so popular?

Paul Lipsky 35:44 Yeah, um, good question. You know, this NotebookLM, I first heard about it last year, maybe fall of 2025, and I was skeptical of it at first. I was like, "Why do we need this when we already have Gemini," right? Seems to do a lot of the same stuff. But then I used it, and I was like, "Oh my gosh, this is really cool." It's not only an incredibly useful tool, but it is so fun to use, and I think that's what really inspired me, was not just the usefulness of it because, but it was I just wanna show people this. This is so much fun to use. It is so cool. So that's where that video came from, basically.

Calum Johnson 36:27 Yeah. You know, if, if, if someone wanted to... And, and let's just get into it, um, 'cause I love your reaction to it. If someone wanted to have that feeling of like, "Wow, this is so cool" What's the first thing that you would have them do when they open NotebookLM?

Paul Lipsky 36:42 Okay. So you need to come into NotebookLM, and right here is this button that says Create New Notebook. I already have a bunch of notebooks in this account, but we're gonna ignore those for a second. You wanna cl- come in here and click on Create Notebook. And the first thing it's going to ask you to do is to add some sources. This is where NotebookLM is incredibly powerful because you're gonna be able to upload or put in here sources that include, like, any files you wanna upload, you can do that. You can add any links to websites or YouTube videos. You can add things from Google Drive, or you can just copy and paste text. And when you do that, you're gonna have all of your sources over here on the left. You can see I have YouTube videos, articles in here, something I've uploaded. And what makes this so unique is that now we have created for us an environment here. We have a chat in the middle, we have this studio over here on the right, I'll show you in a second, and this entire workspace is grounded in these sources. What that means is that when you first add these sources, Gemini, the Gemini inside this notebook, is looking at these website articles and pulling the text from them. It's looking at these YouTube videos and pulling the transcripts from them, and it is creating its knowledge based on that and only on that. So in this case, this notebook is all about how to launch and scale an AI automation agency. I wanted to do some research on that. So there is so much information on the web about how to do that. But when I chat with Gemini here and ask it, like, "How do I get started?" It's not pulling from the web, it's pulling from these trusted sources I have over here on the left. That's what I mean when I say it's grounded, and that is incredibly valuable for a lot of different purposes. A lot of people know that AI can do this and, you know, they have to set it up on their own, but now they don't have to because NotebookLM can do it for them. It's, it's really, really helpful.

Calum Johnson 38:46 Yeah. You know, it-- Let's, let's, let's go deeper there 'cause I think this will be so valuable for someone listening at home, which is let's even take that example, right? Someone's watching this, and they want to build, like, um, an AI automation agency, or they wanna be, like, um, like an AI consultant in the way that you described earlier. I'm curious for you. So that's, like, the overarching topic, and they wanna use NotebookLM to almost become, like, an expert in that. What is your process or, like, the steps that you go through? 'Cause I'm even thinking about how you're choosing-- Like, how you're identifying and then choosing the sources that you're then putting in your notebook. Like, is that just a Google search, and then you're pulling certain websites? Do you have, like, specific places that you go to? Like, just take us behind the scenes of how you're even finding these sources and building out this notebook.

Paul Lipsky 39:42 I'm, I'm smiling to myself 'cause you are asking the right questions that most people don't think to ask, which I love. So when you're creating a new notebook, there are multiple ways you can, you can bring in these sources that I showed you over here on the left. So the sort of lazy way, and I, I do use this lazy way when the topic is not, like, super important. Like, I have a notebook here all about the science of sleep. And so for me, I just wanted some general information about how to sleep better. So for that, when you're creating this new notebook, you can use the research tool right in here. So I can say, "How can I sleep better at night?" And send that along, and it's actually going out and finding sources for me. And then once it finds those, I can approve them or deny them. So that's like-- So you see them all right here. So I can look at all of them, and I can say, "Okay, I like this one." You can even open it up and check the source if you want. And let's say I don't like that one, so I won't check that one off. Won't check that one off, but I like the other ones. And then I can click on Import, and those are being added to my notebook. So that's one way of doing it. But another way is I'll come up here, I'll create a new notebook, and this time you would actually go out and just put in the sources here that you actually trust. So for instance, I have a notebook-- Where is it? Right here, where I grabbed the last maybe 15 of your last or 10 of your last videos, and I put them in here because I wanna be able to chat with this notebook about trending topics on AI. I kinda wanna, like, stay up to date on that. So any-- every time I watch one of your videos now, I throw it in here, and that way, in the future, I can ask, like, "What was that topic they were talking about, um, about Gemini?" And then it'll re- it'll pull that up and say, "Oh, that was the podcast with Paul, and here is a link to it if you wanna check it out. Here's what was said about it." And so with this method, we're only putting in sources that we've already verified, that we already trust. I've already watched these videos. I already know they're great. That's why I added them to this notebook. So it's kind of however you wanna do it. Certainly, this way, where you verify it first and use sources that you actually trust is better. But for some quick stuff, you can just do the search, and it's not a big deal.

Calum Johnson 42:07 Yeah. That's fascinating. So it, it's kinda like two ways. It's, like, one, you can actually-- It's natively built into the tool that it will help identify sources. You can then verify it and add in the ones that you want. Or you could just go and find each one specifically that you want and then only reference those. I think something that's fascinating, though, that I didn't clock the first time, maybe you mentioned it, is, uh, the YouTube videos. So you're actually putting in YouTube videos, and I'm assuming because it's all on Google, because Google owns YouTube, they have, like, the, the transcripts or whatever of these videos.

Paul Lipsky 42:49 Yep. Yeah. So I can actually click on any of these, and you see that here's your video. Here's, um- I think this is a transcript down here, and then up here is a sort of summary of it

Calum Johnson 43:03 Interesting. That's very interesting. So is that-- Is the reason that works... Kind of just a quick question, does that work specifically for YouTube? Or I could also-- I'm assuming it only works for YouTube. I couldn't go to Instagram and put in a link, or I couldn't go to TikTok or LinkedIn. Like, would that also work for those platforms, or is it specifically YouTube?

Paul Lipsky 43:27 You know, to be honest, I'm not 100% certain. I don't want to give wrong information. You, you know, just try it. Throw it in there. [chuckles] It might work. Um, I'm not certain about that 'cause I spend all my time on YouTube, so that's the one that I always use.

Calum Johnson 43:38 Yeah. I'm exactly the same way. I spend all my time on YouTube. It's like the new television.

Paul Lipsky 43:44 Yeah.

Calum Johnson 43:45 Um, okay, great. So I understand the, the data collection piece, and I understand that the, the, like, value prop for Notebook LM is the grounding. It's the fact that you can choose the sources.

Paul Lipsky 44:01 Mm-hmm.

Calum Johnson 44:01 When it comes to actually, like, digesting and consuming that information, can you speak about that part of it? Because there's cool things that you can do, right, with, like, PDFs or it can teach you information or it can talk it out loud. Like, can you explain what's possible there?

Paul Lipsky 44:20 Yeah, absolutely. And I would say that's the, the second value prop of it, is how you can then use that information. The, uh, the grounding's the first one, like you said. So there's two sections here that we haven't covered. Well, we, we covered this one a little bit. This is the chat. So here you can now chat with that information. So here I asked it, "How do I get started?" Super vague question. If I asked Gemini, "How do I get started?" It would say, "How do you get started with what?" But this is, you know, looking at all these sources and it knows I wanna start an AI automation agency. So it gives me a whole breakdown here based on these sources of how to start an AI agency. And you can go back and forth with it asking more questions like, "Help me draft a personalized cold email script," and, you know, just chat with it. Same way you would Gemini, but it's super focused on this topic. But where I think it really shines is inside of Studio over here on the right. We have currently nine different buttons for generating these different assets, and these are really cool. I don't know how else to say it. They're just really cool and really fun to play with. So right here at the top is one of their most talked about standout features. We have these audio overviews. I'm just gonna open up the customizer for it. So when you generate one of these, it actually generates what sounds like a podcast between two hosts, and it's all AI generated. So I've made one of-- I made a bunch of these beforehand because all of these assets admittedly do take a little bit to generate. So we don't waste any time. I just generated one here if you wanna give it a quick listen.

Speaker 45:58 So, uh, what if I told you that starting a profitable AI agency today, like-

Paul Lipsky 46:04 Um, so yeah, it's, it sounds super realistic. It's like two people talking to each other. I mean, that's pretty amazing. And you know, I start to think about, like, the possibilities of this, right? I think the more you play with Notebook LM, the more ideas you're gonna get. Because what if you now create one of these podcast episodes and you actually use that somewhere? Maybe put it on a YouTube video. Maybe use it for marketing your business somehow. I think there's a lot you can do with these that most people surface level aren't seeing.

Calum Johnson 46:38 You know what, Paul? Given that you're, um, I think about what you do and given that you're a, a business owner, what-- 'cause you, you, you marked out like there's nine of these different ways, uh, in the studio. What have you found that you've been using the most frequently? Which ones have been the most valuable to you?

Paul Lipsky 46:57 Sure. Okay. So for any sort of thing that I'm researching, like this notebook about the science of sleep, I love these audio overviews. I generate these and now I have the Notebook LM app on my phone, so when I'm driving, I just play it and now I'm learning about a new topic. It makes it so easy to learn these new things. I don't have to be sitting here in front of my computer to do that. It's really cool. One of my other favorite ones are the video overviews. These are beautiful. So there's currently three different versions of them. Let me show you this one. This is called an explainer video.

Speaker 47:37 Welcome to The Explainer. Okay, today we're unpacking something absolutely huge, the ultimate blueprint for 2026. We're talking about going from zero to owning a highly profitable, low overhead AI automation agency. If you've been paying even a little bit of attention to the business world lately, you know the AI revolution isn't just coming-

Calum Johnson 47:55 That's cool

Speaker 47:55 ... it's already here.

Paul Lipsky 47:57 Yeah, so like super well done. Um, so that's The Explainer, but the new one that I'm really excited for are these shorts. Yeah. Should I play this one?

Calum Johnson 48:11 Yeah.

Speaker 48:12 Most people view sleep as just passively powering down, but it's actually an intense biological maintenance cycle. Here is exactly how lack of sleep breaks your body and how to fix it. First up, the brain's wash cycle.

Paul Lipsky 48:25 I mean, that's pretty cool. And for something that's completely free, that's pretty wild. And this is another one where I, I, you know, this is brand new and I think people are going to take advantage of this and do some pretty cool stuff with these shorts.

Calum Johnson 48:37 Yeah. That is really cool. That is really cool. You know what? Um, and this is an interesting almost like thought experiment for you because obviously you started, like we mentioned in the beginning, you leave your job, what, about eight years ago? Nine years ago? Eight years ago?

Paul Lipsky 48:57 Yeah. Yeah.

Calum Johnson 48:59 To like building your own business. I'm curious for you, let's just say that we fast forward and when you were actually leaving your job was in 2026, and you had this awareness of these tools. Do you think that the way ... I'm curious how you think, uh, that would've changed things, like just the way that you would've approached it. Um, like is there certain things that you look at even when you're using some of this stuff now and you're like, "Oh yeah, for sure I would've been using that in that, in that process of starting a business," or like getting it off the ground, or even what you chose to eventually do? Like what, what comes to mind?

Paul Lipsky 49:39 There's a lot of things that comes to mind. I just ... My brain just raced through like 20 different things when you said that. But, you know, kinda going back to what I said before, just learn this stuff and then bring it to businesses who don't know how to do it. It's, uh, that, that feels so easy to me, like such low-hanging fruit, 'cause people aren't gonna be, aren't gonna spend the time to figure this out. But then looking more practically at stuff like this, like I think a lot of what you have to do as an entrepreneur is try stuff out and see if it works. There's so many ideas. So I look at these Shorts and I'm wondering to myself, "Wait a minute, what if I create one of these Shorts and then come over to Gemini and use this to create an avatar video of myself talking about the same topic? Can I put those together and suddenly have this pipeline where I'm creating content 20 times faster than a normal human can?" Because I don't actually have to sit in front of the camera. I have Gemini making the videos for me inside of Gemini, and then these explainers inside of NotebookLM. You know, I think that scales up at a rate it just couldn't in the past. Uh, will that work? I'm not 100% sure, but again, that's part of being an entrepreneur, trying it out and, and seeing what happens with it.

Calum Johnson 50:51 Yeah. That is such an awesome idea. 'Cause I even think about one of the things I've seen on my Instagram that's performing really well is when you have, uh, the video. You almost have like the split screen with the video of yourself talking about something, and then like the green screen, like an some sort of animation, like visual B-roll, uh, on the top section. And you could see how you could edit those together using that workflow you just described, right? Where you have the avatar of yourself in the bottom section, and then like the animation playing on the top. And to your point, because of AI, you could create a ton of those in a day even. Um, yeah, that's pretty fascinating.

Paul Lipsky 51:31 Yeah. Or even if it's not like ... Even if you don't wanna use an AI avatar, well, you can use it for proof of concept, right? Before you sit down and spend the time making it yourself, make it with AI. And if you look at it and say, "Oh, this wouldn't work," now you didn't waste your time recording it.

Calum Johnson 51:46 Mm. It's such a good point. You know what? If we even rewind a bit deeper into your days when you were working like a 9-5, is there certain ... I'm sure, to be honest, with NotebookLM it feels like there's a ton, but I'm curious what use cases you would even see, like for someone that's working in a corporate job or like in a 9-5. Like are there certain things that you see, like workflows, uh, that you're like, "Okay, this could actually be ... Like I would use this in my day-to-day"?

Paul Lipsky 52:16 Yeah. All right. So you ready for NotebookLM level two? So this is a notebook I created for a, a fake company. It's called Trailhead Supply Company. And instead of uploading websites and YouTube videos as sources, I uploaded company, uh, reports. I have a CSV file that I uploaded with our ad spend, how our, our ads are performing, you know, monthly income and revenue and all that sort of stuff. So all of that's all I uploaded. That's it. And then from that, I asked it questions like ... What's the first question I an- asked? "Rank all five marketing channels from best to worst." And it used that data and gave me that analysis right here. Then I created over here this slide deck, and you can see that this is our plan, our, our ranking. Kind of like a, an overview of how the business is doing and where growth is strong, generated by NotebookLM with just a couple of clicks. And it's a really nice looking slide deck. We have ... What else we have here? What is this one? Oh, here's an infographic. So it also generates infographics s- showing the same information, a kind of audit of the growth of the company and the, the ad spend and how that's affecting it. And then finally we have this report that it put together, 'cause it can generate reports. So if you kinda use this for your own personal data or your business data, it unlocks a lot of possibilities.

Calum Johnson 53:49 Yeah. That's fascinating. That's so good. 'Cause you can actually ... I, I didn't even think about that point. You can upload your own documents, reports, and then you could just build things off the back of that. And so I think about what you used to do being like a lawyer, where you probably have all of these documents, court cases, like legal stuff. You could put it in and then use that to like generate even like a presentation, or even what consultants do. Like strategy consultants, where a lot of the times it's like you're reading through reports or there's like data and then you wanna create slide decks that you can then present. Like it's so ... Even like pitch decks for investors. Like the ... It goes on and on, doesn't it?

Paul Lipsky 54:31 It does. And it's really good at creating slide decks. That's one of my favorite features. Like this is one that it generated. It generates in all these different styles that are just really fun to play around with. Really great infographics. It's, that's one of the places it really shines.

Calum Johnson 54:46 Yeah. You know what, Paul? Before we move on to, uh, Google AI Studios, is there anything we're missing? Is there anything that you, you ... Anything else that you wanna mention in regard to NotebookLM?

Paul Lipsky 54:57 You just gotta play around with it. You're gonna have so much fun clicking these buttons, [laughs] you know, generating all of these assets. Inside of infographics here, for instance, if you click on this arrow, that's how you customize them. And check this out. You can make infographics in all these different styles. You can make it professional if you want, or you can keep it fun. And so just play around with this. You can do different aspect ratios, because the more you use this, the more use cases you're gonna find, and the more you're gonna realize that using AI is kind of fun.

Calum Johnson 55:27 Yeah. Yeah, no, I need to-- I, I'm gonna get on this afterwards. Um, okay, Paul, the, the, the final tool that we, we wanted to talk about is actually the tool that I hear the least about online, and that's Google AI Studios. And so can you just kind of start from the beginning? We've shown people Gemini, which is in, like, G Suite, and really, if you're interacting with Google, even if you're just searching on Google, you're using Gemini. We've shown NotebookLM, which is almost like this ultimate researching tool. Google AI Studios, like, how does it fit into the pie? Like, what is it that it does?

Paul Lipsky 56:10 Well, like I mentioned in the beginning, kind of like to bring this all the way back around, the cool thing about Google is that, and Google AI, is that you can start simple, you know, just using Gemini and using the basic features. You can level up in Gemini. You can use NotebookLM for simple things, or you can level up with it. But then once you're ready to sort of like take that next step and you're like, "I wanna, like, tinker more with AI," that's what Google AI Studio is. So if that sort of interests you and, like, playing around, this is kind of the tool for you. So it's all the same stuff you can do in Gemini, but what's nice about Gemini is that you just ask it to do something. You can change the model here, but there's really only three choices. But it just works. If you wanna kinda dial it in, that's when you can come in here and choose the exact model that you want, and even jump in and give it specific custom system instructions. So it's a little bit more technical on that end.

Calum Johnson 57:09 Yeah. You, you mention it being more, more technical, more custom. Even when I was, uh, reading and doing some of the research, uh, in anticipation of this conversation, it was, uh, saying that initially Google AI Studios was meant really for developers. It feels like, speaking to you today, the everyday person could actually use it. From your experience, is that the case? And then if so, how would you imagine that an everyday person would use Google AI Studios?

Paul Lipsky 57:44 Yeah. So I think it definitely is the case. Um, I think this is one of the more accessible tools that started as a developer tool. And I think in particular, if you're gonna be using this, it's because you're gonna play around with Build. So with Build, you can actually create these apps. And I'll show you some examples of some of the apps, very simple ones that I've, I've created. So I just came in here. This is, you know, if you've ever heard the term vibe coding, it's where you come in here and just literally describe what app you wanna make in plain English, and it will make it for you. So here with just a simple prompt, I generated this app, which is a product description generator. You put in the name of the product that you're selling, so it might be ANC headphones, for instance. You click on Generate description. This one's broken, [chuckles] which is okay because it's easy to fix. Um, you can just tell it. So this was a bad example, so I apologize for that. But the cool thing about this is that whatever just happened, I don't even know what that means. So let me get, let me show you it again. So it was like chair, generate. I get this error message, right? I don't, I don't know what happened here because I'm not a coder. I've never written a single line of code in my life. But what's cool about this is that I can now come over to Gemini and just say, paste in the error and say, "This error came up," and it's gonna fix it for me. I don't know what it's doing, but it's fixing it for me, and that's what's really cool about this. So here's another one. It's just like a simple expense tracking app. And here's another one. Oh, this one's still loading. So yeah, just kind of give you an idea of the, kind of the stuff you can make with it. But I think for, for some people, this may still feel too advanced, so I just wanna give people a heads-up that you can actually do this inside of Gemini. Um, so I've created apps in Gemini. This is Gemini over here. I created this app where you can put in any destination. So I could put in like the pyramids of Egypt, for instance. What are we doing? Egypt. And then you can, like, put in the vibe. So I want it to be like epic adventure. The, I want it to be at midnight. And then I click on this, and it's actually using AI inside the app that I've generated. So there's two places you can generate these. Um, the nice thing about creating them inside of Google AI Studio, though, is that it makes it easier to actually deploy them if you wanna get serious and actually, like, launch this as a real product. Um, just because it connects eas- more easily with the tools for actually deploying it. So if we come back over to this one where we generated the image, in order to share it, there's a couple of options inside of Gemini for sharing apps. If you just click Share, you get a link, and you can share this link with anyone, and they can, uh, instantly access it. But it's not behind a, a paywall. They don't need to give you your e- their email address, so it's, it's like there's, uh, it's, you can't monetize this, right? The other option is you can-- I think I just accidentally closed it. But the other option is you can copy the code, and then you have to bring that code over and actually use it somewhere else. But with Google AI Studio, you can build it directly in here, and then you can send it over, I think, to GitHub directly, for instance.

Calum Johnson 1:01:13 Yeah. Okay, that's, that's fascinating. It's like, I think one of the things that's been, uh, crazy to me just in the last year is the ability that anyone listening at home actually has to, like, build things, like build software and apps. And to your point, it's like you haven't written a single line of code in your life. And even when the thing breaks, it's like you just tell it that it's broken and it will fix it. I just think it's, uh, it's like this insane ability that we have in 2026.

Paul Lipsky 1:01:50 Yeah, it really is. And if you start to think about this, and, you know, when you first start using this, your apps aren't gonna be very good. They're gonna be like this, like very basic. Um, and then the more you use it, the more you're gonna so- start to find, like, cool use cases for it, like how you can build out entire websites, or you can generate apps like this which actually have AI built in. And that starts to get you thinking, "Okay, I'm a business owner. What tools am I paying for that I can just Vibe code an app for free and then use forever for free so I don't have to pay for those apps anymore?" And then, and then it's like, wait a minute, can I do this for other businesses? And if you learn how to Vibe code decent internal tools and to get them, in a way, accessible to businesses, they're willing to pay money, especially if it's a lot cheaper than some of the enterprise products out there, which can be extraordinarily expensive.

Calum Johnson 1:02:47 Yeah. You know what, Paul, just on that, on that point of even for business owners, that as you get more familiar with, uh, like these tools like Google AI Studio and start building apps, what to you have you found, even in your business and as you've been using this technology, what kind of tools, apps have you built that have actually been valuable, useful? Like it's gone above just the playground. It's actually like, "Okay, I actually-- This has a use case for me."

Paul Lipsky 1:03:22 Yeah. So one of the apps I built will generate, um, well, can, you can choose how many it generates, generates multiple thumbnails. So that uses NanoBanana, same way that this app is, how it's using NanoBanana to actually generate the image that appears in here. It's using NanoBanana to generate six thumbnails I usually have it make for me in different styles so I can look at them and then choose which one I wanna use for my videos and then iterate on it. So something like that is pretty cool. So, you know, I can, I can come into Gemini and say, "Create a thumbnail for me, create a second, create a third, fourth, fifth, sixth," but the app just makes it easier 'cause it'll just generate all six at once. So that's been a, a pretty nice one. Um, and there are tools that you can pay for that do that, but why would you when you could just have an app that you make? It took me probably five minutes to make that app.

Calum Johnson 1:04:11 Yeah. You know the thing that's like, even as you're describing it, it's clicking for me? Part of the advantage that Google has is they have so many of these different tools, and so it's like when you wanna do image generation, you're pulling from NanoBanana or it's using Gemini or it-- Like there's so many things that it has but all in one. And then you mix that in with also the G Suite access, right? Which is like, oh yeah, I wanted to create a PowerPoint in Slides, or I wanted to create spreadsheets in Sheets, or I wanted to write a memo in Google Documents. So it's like the, the, just the, the breadth of tools that it has all in one feels like kind of like a unique value proposition for Google specifically.

Paul Lipsky 1:04:59 It really is. And part of the reason is because they have so much money and so many resources, and they've always believed-- They have this, they have this program called Google Labs, and the whole point of Google Labs is to try things. And that's how Notebook LM came about, because someone just had this idea for creating this system where you can create these notebooks that are grounded, and it just started as a little experiment and it took off. People absolutely loved it. Some of the ideas they create with Labs go nowhere, and others get, like, absorbed into other products. So it might start out as just, like, a single app that does something very specific, and then they bring that into Gemini. That's how Music started, for instance, and now it's just part of Gemini. And so because of that, they're just always releasing these new tools that do all this crazy stuff. So this is another tool called Flow. This is specifically for image and video generation. Again, you can do this inside of Gemini, but this is really nice because you can have this whole workspace where you have all the images you're working on, all the videos. Here's another one that they do, which is just for music generation. Um, so just tons of tools that they have for different purposes. And the cool thing is that I know it can feel overwhelming because I think they have, like, over 30 of these, but you don't have to use them all, right? You just have to know the, the main important ones, which I think are Gemini and Notebook LM. And then if you're ready for the extra tinkering, you can go into AI Studio or try out some of these other ones.

Calum Johnson 1:06:32 Yeah. You know, one of the things is, and, and, and you'll resonate with this, Paul, I think one of the, the principles in, in business that you learn pretty early is that of, like, the zig when they zag, is doing the thing that not everyone else is doing or using the thing that not everyone else is using. And I think back to what we were talking about earlier, where everyone, for the most part, is on either ChatGPT or Claude. For you, just given the, like, expertise and really just the time that you've spent with Google's AI tools, does it feel like you have an edge? 'Cause a, a lot of what you've shown me today I didn't even really know was possible. Specific- especially with, uh, Notebook LM, I didn't even really know that you could do that, and I'm using Claude and ChatGPT every day.

Paul Lipsky 1:07:25 Yeah, um, you mean an edge in terms of, like, my content creation or just in business in general?

Calum Johnson 1:07:31 Just that the fact that you, you kind of even in the way that you were using it in your business, your content, even in your life with the sleep use cases, it feels like it kind of has given you something, um, that even the other tools haven't.

Paul Lipsky 1:07:46 Yeah. Yeah, I do feel that way. I think, I think that a lot of people sleep on what, uh, no pun intended, on what you, what's available to you through Google. Um, and people would just, like, get hyper-focused on ChatGPT or Claude without realizing that these incredible tools exist. And, like, look at this one, for instance, this, this music generation. I mean, how many people know about this? Very few, and this is, this is a very good music generation, uh, AI tool. And so, you know, there are possibilities here. What can you do with AI music? You can go out, you know, figure that out. But not everyone's gonna know about this, so they're not gonna know how to do this. And if you can be, like, one of the people figuring it out, that gives you an advantage, especially when people don't even know it exists.

Calum Johnson 1:08:31 Mm-hmm.

Calum Johnson 1:08:32 You know what, Paul? Because I want to, um ... You mentioned some of the apps, and I think that the, the vibe coding and the ability to build anything will be one of the most valuable takeaways that someone would take from watching this conversation. For some of those apps that you've built, because it really starts with, like, the prompt and what, you know, what you instructed the AI to do. Can you just walk people through from, like, the step one how you even thought about prompting, describing, outlining to Google AI Studio what exactly you needed so that it could then build the app that you wanted? 'Cause I can see that being a frustrating experience for people is, like, they think they're being clear, but they're not getting the output that they want.

Paul Lipsky 1:09:25 Yeah, yeah, yeah. Good question. So I'll show you this example 'cause I already have these ones pulled up inside of Gemini. Um, you know, the term is called vibe coding for a reason, and it's because what you're describing to the AI is the vibe. You, of course, wanna explain what you want the app to do. So this is an app that's, that's just a game. So it's, like, an arcade where you can actually play three different games in here, and you can just describe the, what you want it to do. So I w- said I wanted a arcade style system with three different games, but then you want to describe the vibe of it. Like, I want it to feel very retro. I want it, like, someone playing it to feel like they're back in the '80s playing this game on, like, an old system. Um, so that's, I think, very important to describe those vibes. And then I also think it's important to give it, um, what's the word I'm looking for? Uh, examples. So if you see something out there that you like, like a style. So if someone else posted this and you're like, "I really like that style," take a screenshot of it and give it to Google AI Studio or Gemini and say, "I really like this style. I don't wanna create this same app, but I wanna use this style in this other app that does this other thing." It's really good at looking at references like that when building out what you need. But the other important thing that people get wrong when it comes to vibe coding these different apps is that they try to give it too much at once, and this is a little bit of an ex- Actually, there's a better example of it over here where I created this sort of note-taking system, and there's just too much going on and for the first, for the first run. The first thing you should do is sort of like the, the minimal viable product. I'm wanting to create a note-taking system, so instead of doing all this stuff, I should have just had it create for me this, where I just create notes in here and that's it. And then once I verified that worked and I like the aesthetic of it, then we can iterate on it and add an inbox and the interactive graph and the daily AI synthesis. Because if you try to do too much at once, something's gonna break. You try to fix that one thing, then you realize something else doesn't work or it breaks something else and, and it can get frustrating. So starting small and having, like, incremental changes is a really smart way to get the app to work every step of the way, and so the final product is much better.

Calum Johnson 1:11:50 Yeah. You know, if I was gonna distill it, and, and you let me know if I'm correct in this, Paul. It sounds like there's three parts to, like, the perfect prompt in terms of building apps here. So it starts with, like, what is it that you want it to do? And you focus on, like, one core use case, which in this case is, like, I just want it to be a note taker or I just want it to be, like, a, a note ... It gathers notes. Like, that's the main use case, what it does. Then you ... The second part is you have, uh, the vibe. Like, here's how I want it to feel. Here's how it kind of looks. Here's kind of, like, the... And then, and then that moves us on to the third part, which is the style, which is where you're adding, like, screenshots of inspiration so it can get the exact look. Um, you're using words like, "I want this retro style," or, like, uh, like, "this minimalist," like, very simplistic style. Cartoon style I could imagine as well. That seems like kind of the three-part process. It's like, what does it do? And then vi- what's the vibe of it? Then what is the style? And that will kind of give you your first, uh, like, great output.

Paul Lipsky 1:12:57 Yeah, exactly. I think you nailed it.

Calum Johnson 1:12:59 Sweet. Nice. You know what, um, Paul, once, uh, before we get out of here, and, and it's funny 'cause you and I were actually, um, we were speaking about this before, which I think when it comes to videos like this, it's so easy to watch it and get inspired and come up with all of these ideas, and then the action never happens. And so I'm curious for you f- from your vantage point as someone that made that- Both you made that transition from like working a 9-5 to now being online and having your own business. For someone that's listening to this conversation and they enjoyed this conversation, what would you have as like the immediate step one, like the thing that they should go and do immediately after listening to this conversation so that they can take action?

Paul Lipsky 1:13:58 Yeah. It's really not that hard, my, my tip for you, and that's just to go out and start playing with these tools. Have fun with them, seeing what they can do. I would actually start with NotebookLM, especially if you haven't used it already, because you can come in here and just clicking on all these different buttons on the right is just a lot of fun. You're gonna start to get an idea of what you can make with this, the different types of in- infographics and the different styles. Because the value that you're going to have is understanding what these tools can do when other people don't, and the only way you're gonna get that experience is just coming in here and playing with them.

Calum Johnson 1:14:38 Yeah. You know, you even, um, you said this and I wanted you to elaborate on it 'cause I think it's so fricking good. You said, "I always encourage people to just try to do everything with AI just for three days. Take everything that you normally do on the computer, then try and have AI do it." Can, can you just share with people why did you say that?

Paul Lipsky 1:15:04 Yeah. It, it's a really good exercise in forcing yourself to see what the limits are of the tools that you use. Because I think so many times people are stuck with that question that we raised before, what can you do with AI? What are the workflows? And a lot of people don't know that AI is capable of much more than they think if they just spend the time to learn how to do it. So the first time that ... Okay, let's say it's email. The first time you come into Gemini and you're like, "Can Gemini help me with my emails? I don't know." You come in here and you ask Gemini, "Can you help me with my emails?" It might say no. Okay, now you know. But then you ask it, "How can AI help me with my emails?" And then it'll probably tell you, "Go into Gmail," and it will actually have a, a message, a, the box down there where you can have it write for you. Or an, you know, this was a silly example because Gemini can help you with your emails, but that's what it'll tell you. So it's all about just trying to, trying really hard. And the first time that you do it, it might take you a long time because it feels clunky 'cause you're not familiar with these tools. But then the more you use it, you're like, "Hey, this is speeding up my workflows. I'm able to get through my emails twice as fast as I used to because now I understand how that works." And now you might be thinking, "Okay, what if I teach this to other people? What if I go into businesses and help them streamline the processes they have using AI?" Well, that's a valuable skill, and it's all because you took the time to learn how to do it, um, when no one else was taking that time.

Calum Johnson 1:16:45 Yeah. You know what, to your point, Paul, I think it's like a, it's really like a mindset switch because we're so used to just doing things. Like we're so used to manually doing all of the things ourselves of like writing the emails or writing the memo or gathering the data and putting it in the spreadsheet or building the slides manually ourselves. It's actually like a mindset switch to be like, "Oh, no, the technology can actually do it for me in a very custom way." And it, and, and where my mind goes, I think about what you mentioned of like saving the time. And so can you even just speak to that portion for you which is like the value in your life, in like your personal experience of being able to use AI in the way that you are? Like what is it that you've seen? I even think about the freedom that you mentioned in the beginning.

Paul Lipsky 1:17:41 Mm-hmm. Yeah. So AI touches everything I do now because I, I did that experiment and I do it every few weeks because the models and the tools are getting better and better. And so now my, I wake up and I get a report from Gemini every morning. See it over here on the left, it says Daily Brief. Again, unfortunately I can't click on this 'cause it'll actually like say what's on my calendar. You know, I have personal stuff in my calendar, personal emails. But this will give me a breakdown of my entire day and let me know what's going on for the day. So mentally I'm in the right space. I go into Gmail. I already have drafts of emails in there. I can quickly reply to everything. It's auto sorting them. Um, I, you know, have research I have for my videos and the outlines. I have Google Drive outline those videos for me. So it's just like such a time saver. You know, it's not replacing me. I'm still in the loop every step of the way. It's not sending out emails for me, but it's speeding up the things that I already do. But the only reason it's doing that is because I asked it, "Can you do this? How can we do this faster?" And I figured those things out.

Calum Johnson 1:18:51 Yeah. You know what, b- Paul, before we get out of here, when you think to the, to the future, and I'm just curious from your vantage point, given the time that you spend with this technology, the people that you speak to, um, even the people that reach out to you that watch your videos, where do you think the future of this is going? Because it, it can already do so much and I, and I'm curious to know, is there something where you're like, "I actually think that a year from now, two years from now, we're gonna be in this position"? Like kinda share that future or what, what you see.

Paul Lipsky 1:19:30 There's so many directions I can go with that. [laughs] You know, I think that one of the things that excites me is being able to pick up my phone and just pressing a button and speaking to Gemini or whatever tool it is I wanna use and just saying, "Do this, do that." Speaking to it live saying, "What emails did I get today?" You know, as I'm out for my walk, and it's telling me the emails and it's saying, "I already drafted these replies. Do you want me to send them?" Say yes, and all that's just kind of being done. Like really having that feeling that I have a personal assistant that is human-like to talk to, and I think that is coming. There's already some tools like that in beta that we've seen previews of, and so that's kind of one of the things that really excites me.

Calum Johnson 1:20:17 Yeah. It's like Jarvis from Iron Man.

Paul Lipsky 1:20:20 Exactly. [laughs] Exact- that's what I want.

Calum Johnson 1:20:22 Yeah. That's awesome. Paul, let people know like where they can, uh, where they can find you.

Paul Lipsky 1:20:29 I'm all in on YouTube, so if you wanna connect with me, just check out my YouTube channel. It's just my name, Paul J Lipsky. Subscribe, leave a comment. I reply to every single comment. Um, so yeah, that's the best place to find me.

Calum Johnson 1:20:43 Paul, you've been amazing, mate. This has been awesome.

Paul Lipsky 1:20:46 Yeah, it's been a lot of fun. Thank you for having me on.

Calum Johnson 1:20:48 So if you enjoyed this conversation and you wanna hear even more stories like this, then just click here. And also my team is gonna put some more videos that you can watch here. Thank you.